I am looking for a circuit to increase input DC voltage in a 20%. The input range will be 10 VDC to 15 VDC. The output should be 12 VDC to 18 VDC. Can you help me?
I’m assuming the input is DC. How much current do you need?
The simplest solution is likely a MC34063 with an inductor (about 20 uH) and a timing capacitor (about 470 pF) and a Schottky diode (capable of a few amps) and a buffer capacitor (typically 470 uF) as well as a resistor divider (27 kOhm and 2 kOhm resistors.) Hook it up in boost configuration, at 50 kHz. This will actually regulate the voltage to about 18 V.
If you need more than the rated 1.5 A of inductor current, you can use a N-channel MOSFET and a gate driver like the IR2001 for the switch, instead of the built-in Darlington pair of the 34063.
Now, if there is some special reason why you absolutely can’t use a regulated boost circuit like this, you can turn your DC input into AC by using a PWM generator – a 555 timer circuit with a power MOSFET for switching on/off would work fine. Set it for a 50% duty cycle. Hook up a transformer, with 5:6 winding ratio. Then use a LC filter to smooth out the voltage on the other end. However, you may see voltage sag on the other end when you apply load.
Thanks for your response. I need < 1,5 A.
As I said in my first post, the input is DC. The problem is not to build a regulated circuit with a variable input VDC and a fixed 18 VDC output. I need the output VDC to be about 20% more than the input voltage, so when input is 10 VDC the output should be 12 VDC. If input is 12 VDC, the output should be 14,4 VDC, and so on.
It will be perfect if the percentage of voltage increase can be adjustable, f.i. from 15% to 25 %.
In many years of teaching electronics, I’ve never heard of such a circuit and can’t really imagine why anyone would need it. There is probably another solution to whatever problem you are trying to solve.
What you are asking for is indeed a regulated boost converter, with an output voltage reference that is 1.15-1.25 times the input voltage. The voltage multiplier could be complicated to design but can be done using either analog or digital techniques. Unfortunately all such approaches, to my knowledge, require other regulated voltages to be present in the circuit (and at least one must be higher than the output voltage).
At what frequency do you expect the input voltage to change, and with what frequency response do you expect the output to faithfully reproduce those variations?
Thanks for trying to help me.
For those not related with automotive electronics, Mercedes Benz front SAM is a control unit that gives power to several items in the front part of the car. This front SAM gives power to headlights, fog lights, and DRL lights, for instance.
In Europe, day running lights (multi-LED) work at full brightness under day light (day mode). In the evening (night mode), when low beam is activated, DRL are automatically dimmed. The dimming factor is controlled by the SAM and can be adjusted from 100% to 0% dimming in fractions of 10%. Full brighness is obtained with 14,5 VDC (with engine on, otherwise, DRL are off). For dimming function, SAM can be adjusted to get the desired brightness by decreasing the VDC going to the DRL. For instance, 80% of dimming gives a voltage of 11 VDC; 40% of dimming gives 12,8 VDC from the SAM to DRL.
Ok, my problem is I am trying to use non-original DRL lamps. These DRL LED work with a voltage 20-25% higher than the original, so they are off at 12 VDC, dimmed at 13 VDC and go to full brightness at 17,5 VDC.
The boost circuit I am asking for will work at around 13 VDC or with 17,5-18 VDC depending on the input voltage from the SAM. I can adjust the low voltage by adjusting the dimming factor in the SAM through programming with Xentry tool.
So to use these DRL LED I just need to increase the output voltage from the SAM a 20-25%. With this circuit between SAM and non-original DRL LEDs, the DRL behaviour will be the same as the original, so there will be no need to extra-wiring.
"At what frequency do you expect the input voltage to change, and with what frequency response do you expect the output to faithfully reproduce those variations?"
Once or twice a day? This is not the problem, as you can understand from this post. 
As I imagine you realize, it will be much simpler just to buy the correct lamps.
Sure, but they cost ten times(!) more, so it’s worth to build the circuit.
This is not a simple circuit to design and it won’t be cheap to build. The most straightforward approach would be to modify a microprocessor-controlled power supply circuit. I suggest to post to the Spark Fun forum. Someone there might be willing to help.
I supposed that Pololu DC-DC could be modified for my application needs. Thanks for your suggestion.
Hi guys, whats the verdict? I am also looking for such module…
Using it for an Solar light LED dimmer controller that has output of 24v but the LED only works at 33-40v
I need to step up the dc voltage but then i need the converter output to change linearly according to the converter input so that the LEDs can be dimmed
It is probably most practical to use a different approach, like Jim mentions above, for something like that.
-Derrill